Granted Patent B2

US 12,593,097 · App. 18/016,598 · Granted Mar 31, 2026

Inventors: Félix Henry (Saint-Grégoire, FR); Gordon Clare (Pacé, FR)

Assignee: ORANGE

H04N21/435H04N21/44209

Abstract

Claims (31)

1 . A method for decoding a data stream including an indicator and representative data that is representative of an audio or video content, the method being implemented by a decoding device, the method comprising:

selectively decoding the indicator having one of a first value and a second value different from the first value, the first value indicating that an artificial neural network to be used to decode said representative data is encoded in the data stream, the second value indicating that the artificial neural network to be used to decode the representative data belongs to a predetermined set of artificial neural networks accessible by the decoding device; and

decoding said representative data by the artificial neural network.

2 . The decoding method according to claim 1 , wherein said audio or video content is a first part of a video sequence, said video sequence comprising said first part and a second part,

wherein the decoding said representative data produces said first part, and

wherein the method further comprises decoding other data by another artificial neural network to produce the second part.

3 . The decoding method according to claim 2 , wherein the other artificial neural network has an identical structure to a structure of said artificial neural network.

4 . The decoding method according to claim 3 , wherein the first part and the second part respectively form two groups of images for a format of representation of the content used.

5 . The decoding method according to claim 2 , wherein the first part and the second part respectively form two groups of images for a format of representation of the content used.

6 . The decoding method according to claim 2 , further comprising transmitting a list of artificial neural networks to a controller configured to control data stream transmission.

7 . The decoding method according to claim 1 , further comprising transmitting a list of artificial neural networks to a controller configured to control data stream transmission.

8 . The decoding method according to claim 1 , further comprising, decoding an identifier of the artificial neural network, when it is determined, by decoding the indicator, that the artificial neural network belongs to the predetermined set of artificial neural networks.

9 . The decoding method according to claim 8 , further comprising reading, in a storage, parameters of the artificial neural network identified by the decoded identifier.

10 . The decoding method according to claim 9 , wherein the storage stores a first set of parameters representative of a first artificial neural network forming a random access decoder and a second set of parameters representative of a second artificial neural network forming a low latency decoder.

11 . The decoding method according to claim 8 , further comprising generating an error message in an absence of data related to the artificial neural network identified by the decoded identifier.

12 . The decoding method according to claim 8 , further comprising receiving, from a remote server, parameters of the artificial neural network identified by the decoded identifier.

13 . The decoding method according to claim 1 , further comprising, decoding data encoding the artificial neural network that are included in the data stream in order to obtain parameters of the artificial neural network, when it is determined by decoding the indicator that the artificial neural network is encoded in the data stream.

14 . The decoding method according to claim 1 , wherein the decoding device stores parameters defining the artificial neural networks of the predetermined set of artificial neural networks.

15 . The decoding method according to claim 1 , wherein the decoding device has access to parameters defining the artificial neural networks of the predetermined set of artificial neural networks by connection to remote electronic equipment.

16 . The decoding method according to claim 1 , wherein the indicator allows the decoding device to know the method by which the decoding device will be able to have access to the artificial neural network to be used to decode the representative data.

17 . A decoding device comprising:

a receiver configured to receive a data stream comprising an indicator and representative data that is representative of an audio or video content, the indicator having one of a first value and a second value different from the first value; and

one or more processors configured to decode the indicator, decode the representative data using an artificial neural network encoded in the data stream when the indicator has the first value, and decode the representative data using an artificial neural network belonging to a predetermined set of artificial neural networks accessible by the decoding device when the indicator has the second value.

18 . A method for decoding a data stream including an indicator and representative data that is representative of an audio or video content, the method being implemented by a decoding device and comprising:

decoding the indicator, which has one of a first value and a second value different from the first value, the first value indicating that an artificial neural network to be used to decode said representative data is encoded in the data stream, the second value indicating that the artificial neural network to be used to decode the representative data belongs to a predetermined set of artificial neural networks accessible by the decoding device;

selecting among a first decoding process and a second decoding process based on the decoded indicator,

the second decoding process comprising decoding an identifier of the artificial neural network in order to obtain parameters of the artificial neural network and decoding the representative data using the artificial neural network configured with the parameters obtained according to the second decoding process; and

the first decoding process comprising decoding data encoding the artificial neural network that are included in the data stream in order to obtain parameters of the artificial neural network and decoding the representative data using the artificial neural network configured with the parameters obtained according to the first decoding process; and

based on the selecting,

executing the second decoding process when the decoded indicator has the second value,

else executing the first decoding process when the decoded indicator has the first value.

Assignments (2)

Priority Claims (1)

Continuity (1)

References Cited (22)